#video on #3D-printing an #STM for #microscopy. This is a unimorph disc type made from a piezo buzzer; he ruined several before switching to ultra-fine magnet wires and low-temperature solder (ChipQuik?). His kinematic mount seems to consist of three bearing balls, one in a V-groove, one on a flat glass surface, and one in some kind of a pit, each adjustable with a knurled screw. He seems to have used heat-set inserts to build his frame out of 3-D printed plastic — although PLA has the best TCE of common plastics, he’s using PA-CF, carbon fiber reinforced nylon, which ought to be pretty stable in the X and Y dimensions, like wood. (He mentions that many STMs are made out of Zerodur or similar glass-ceramic #materials.) Printed on a Bambu Lab X1 Carbon. The voiceover might be an AI reading a script because it mispronounces a lot of words and sometimes gets the prosody of sentences wrong, despite having a bog-standard GA accent. He put metal tape on the parts for EMI shielding; looks like aluminum flashing. To pre-amplify the nanoamps or picoamps signal from the tunneling current, he copied Dan Berard’s open-source STM preamp circuit, naturally using a current-to-voltage converter built around an op-amp, but I can’t tell which one. He shielded his cable to the probe and put the analog electronics in a Faraday cage to reduce EMI further, getting only about 10pA of noise. Berard’s #electronics design used an OPA124, but he says “The chips used in Dan’s design are hard to find these days.” When it comes to #microcontrollers #hardware, he’s using a Teensy 4.1. He’s using an aluminum plate and some rare-earth magnets to damp the oscillations of his stable platform. “This is the first atomic image captured by a 3-D printed scanning tunneling microscope.”
on 02024-09-28why #STM #microscopy is usually done at ultra-high #vacuum (picotorrs). Handy chart: roughly XHV = attotorrs, UHV = picotorrs, HV = nanotorrs, MV = microtorrs, rough vacuum = millitorrs. “At a pressure of 1×10⁻⁶ Torr the crystal surface will be contaminated with a monolayer of gas in 1 second.”
on 02021-11-11David Berard’s #STM #hardware (and software) for #nanotech #microscopy in air, using the software Gwyddion. Won’t work on most metals because of the oxide coating.
on 02021-11-10a note on #STM being dropped in C# #transactional
on 02019-01-21how software #transactional memory (#STM) failed in C#
on 02019-01-21